The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation

Simon M. LUCAS; Jialin LIU; Diego PÉREZ-LIÉBANA

doi:10.1109/CEC.2018.8477869

The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation

Simon M. LUCAS, Jialin LIU, Diego PÉREZ-LIÉBANA

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Referred Conference Paper › peer-review

35 Citations (Scopus)

Abstract

This paper describes the N-Tuple Bandit Evolutionary Algorithm (NTBEA), an optimisation algorithm developed for noisy and expensive discrete (combinatorial) optimisation problems. The algorithm is applied to two game-based hyperparameter optimisation problems. The N-Tuple system directly models the statistics, approximating the fitness and number of evaluations of each modelled combination of parameters. The model is simple, efficient and informative. Results show that the NTBEA significantly outperforms grid search and an estimation of distribution algorithm.

Original language	English
Title of host publication	2018 IEEE Congress on Evolutionary Computation, CEC 2018 : Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Number of pages	9
ISBN (Electronic)	9781509060177
DOIs	https://doi.org/10.1109/CEC.2018.8477869
Publication status	Published - 2018
Externally published	Yes
Event	2018 IEEE Congress on Evolutionary Computation, CEC 2018 - Rio de Janeiro, Brazil Duration: 8 Jul 2018 → 13 Jul 2018

Conference

Conference	2018 IEEE Congress on Evolutionary Computation, CEC 2018
Country/Territory	Brazil
City	Rio de Janeiro
Period	8/07/18 → 13/07/18

Bibliographical note

Publisher Copyright:
© 2018 IEEE.

Keywords

Estimation of Distribution Algorithm
Evolutionary Algorithm
Game Playing Agent
Hyper-Parameter Optimisation
Noisy Optimisation
Rolling Horizon Evolution

Access to Document

10.1109/CEC.2018.8477869

Cite this

@inproceedings{8cb26add5a204d7f8b1d7ba92c69051e,

title = "The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation",

abstract = "This paper describes the N-Tuple Bandit Evolutionary Algorithm (NTBEA), an optimisation algorithm developed for noisy and expensive discrete (combinatorial) optimisation problems. The algorithm is applied to two game-based hyperparameter optimisation problems. The N-Tuple system directly models the statistics, approximating the fitness and number of evaluations of each modelled combination of parameters. The model is simple, efficient and informative. Results show that the NTBEA significantly outperforms grid search and an estimation of distribution algorithm.",

keywords = "Estimation of Distribution Algorithm, Evolutionary Algorithm, Game Playing Agent, Hyper-Parameter Optimisation, Noisy Optimisation, Rolling Horizon Evolution",

author = "LUCAS, {Simon M.} and Jialin LIU and Diego P{\'E}REZ-LI{\'E}BANA",

note = "Publisher Copyright: {\textcopyright} 2018 IEEE.; 2018 IEEE Congress on Evolutionary Computation, CEC 2018 ; Conference date: 08-07-2018 Through 13-07-2018",

year = "2018",

doi = "10.1109/CEC.2018.8477869",

language = "English",

booktitle = "2018 IEEE Congress on Evolutionary Computation, CEC 2018 : Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

}

LUCAS, SM, LIU, J & PÉREZ-LIÉBANA, D 2018, The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation. in 2018 IEEE Congress on Evolutionary Computation, CEC 2018 : Proceedings., 8477869, Institute of Electrical and Electronics Engineers Inc., 2018 IEEE Congress on Evolutionary Computation, CEC 2018, Rio de Janeiro, Brazil, 8/07/18. https://doi.org/10.1109/CEC.2018.8477869

The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation. / LUCAS, Simon M.; LIU, Jialin; PÉREZ-LIÉBANA, Diego.
2018 IEEE Congress on Evolutionary Computation, CEC 2018 : Proceedings. Institute of Electrical and Electronics Engineers Inc., 2018. 8477869.

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Referred Conference Paper › peer-review

TY - GEN

T1 - The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation

AU - LUCAS, Simon M.

AU - LIU, Jialin

AU - PÉREZ-LIÉBANA, Diego

PY - 2018

Y1 - 2018

N2 - This paper describes the N-Tuple Bandit Evolutionary Algorithm (NTBEA), an optimisation algorithm developed for noisy and expensive discrete (combinatorial) optimisation problems. The algorithm is applied to two game-based hyperparameter optimisation problems. The N-Tuple system directly models the statistics, approximating the fitness and number of evaluations of each modelled combination of parameters. The model is simple, efficient and informative. Results show that the NTBEA significantly outperforms grid search and an estimation of distribution algorithm.

AB - This paper describes the N-Tuple Bandit Evolutionary Algorithm (NTBEA), an optimisation algorithm developed for noisy and expensive discrete (combinatorial) optimisation problems. The algorithm is applied to two game-based hyperparameter optimisation problems. The N-Tuple system directly models the statistics, approximating the fitness and number of evaluations of each modelled combination of parameters. The model is simple, efficient and informative. Results show that the NTBEA significantly outperforms grid search and an estimation of distribution algorithm.

KW - Estimation of Distribution Algorithm

KW - Evolutionary Algorithm

KW - Game Playing Agent

KW - Hyper-Parameter Optimisation

KW - Noisy Optimisation

KW - Rolling Horizon Evolution

UR - http://www.scopus.com/inward/record.url?scp=85056256593&partnerID=8YFLogxK

U2 - 10.1109/CEC.2018.8477869

DO - 10.1109/CEC.2018.8477869

M3 - Conference paper (refereed)

AN - SCOPUS:85056256593

BT - 2018 IEEE Congress on Evolutionary Computation, CEC 2018 : Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 IEEE Congress on Evolutionary Computation, CEC 2018

Y2 - 8 July 2018 through 13 July 2018

ER -

The N-Tuple Bandit Evolutionary Algorithm for Game Agent Optimisation

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this